David Rousseau and I have both written books about the relationship between war and rights (or, in my case, war and liberty more broadly). His is descriptive; mine, prescriptive. He aims to describe the impact that war has on rights at home; I make an argument about what the relationship between war and ordered liberty ought to be at home and abroad.

Rousseau outlines three possibilities: that war hurts rights; that war helps rights; or that the relationship is complex and changes over time, hurting in the short term and helping over the long run. Drawing on the relevant body of literature, he calls notions that war hurts rights the “Garrison State” hypothesis: the national security state cracks down on dissent in wartime to ensure national cohesion and prevent sedition. The idea that war helps rights he calls the “Extraction” hypothesis: war is expensive, and states must buy citizens’ loyalty and sacrifice by conceding to their demands for more say in how things are run.

Which is true? He argues, like all good scholars, that it depends. Specifically, it depends on how big the threat is and how involved is the population. A small threat with an uninvolved population yields no change in rights at all. (A small war with a highly invested population is a null set and plays no role in his analysis.) A big threat with an uninvolved population results in the Garrison State. The state takes the war seriously. but the population does not have a stake in it, and so the people lack leverage to demand accountability. A big threat with large popular participation is where things get interesting. In that case, the state starts out acting like a Garrison, but as the war wears on, the population demands reform as part of the bargain for their wartime sacrifice. In this case, rights follow a “J-curve,” declining early on before improving in the long run. Rousseau bolsters his case with large-N statistical work and historical case studies drawn from different countries and time periods, relying on the mixed-methodology approach currently favored in political science. His cases on the Austro-Hungarian Empire and, later, on African Americans’ military service are especially interesting. If there is a flaw in the book—and this is a quibble more than a flaw—it is that some of the statistical material will be inaccessible even to other scholars and might have been better relegated to an appendix, leaving the main text to focus on the argument. His choice of case studies is also a double-edged sword. Although the diversity of cases bolsters the scope of his claims, one might ask whether the cases can be too diverse; for example, is the case of ethnic minorities in Imperial Russia comparable to the global progress of women’s suffrage in the twentieth century? But I think the first cut of the blade is sharper: his study has an admirable ambition and sweep to it, which require a broad case selection.

The chapter on African Americans’ military service and their gain in civil and political rights is an excellent example of how a mixed-methodology research design generates a strong argument. Rousseau surveys the history of African Americans’ military service in major wars from the American Revolution through World War I, showing how each war was followed by incremental gains in their legal and political status in at least part of the United States. He marshals statistical data, where available, to demonstrate the level of mobilization. He then spends the bulk of his chapter on World War II, for which survey data from African American soldiers are available. The survey data enable Rousseau to explore his hypothesized causal mechanism: that high mobilization triggers a bargaining process in which the citizenry exchanges wartime support for more rights at home. The data strongly support Rousseau’s main contention.

If Rousseau is right—and his argument is persuasive on its face—I conclude that the United States and other democracies ought to reinstitute the draft. There is nothing to be gained by an uninvolved population, regardless of the size of the threat. If one believes, as I argue in my book, that war ought to aim at the vindication of ordered liberty at home and abroad as the only legitimate just cause and just effect of war; and if, as Rousseau argues, wars end up restricting liberty if populations are uninvolved in them; then to safeguard liberty in wartime, ensuring a high degree of popular participation is essential.

I suspect it is also likely that a more involved population would be more selective about the wars it fights and more likely to demand either rapid withdrawal or complete victory. It is highly likely that there would have been substantially more, and earlier, opposition to the war in Iraq; conversely, the population might have felt far more invested in the war once it started, in turn empowering them to demand more rapid adaptation and accountability when the administration’s strategy proved faulty, instead of waiting until 2007. Either course had risks and benefits, but they both had strategic logic to them. A highly involved population might have debated the merits of either choice and, either way, owned and borne the risks together. Instead, the US population was disengaged compared to prior US conflicts, enabling US policy makers to choose the worst of all possible strategies: they tried to split the difference between going all in, on the one hand, and heading home, on the other.

From Rousseau’s argument, we would conclude that either case—using force massively or not at all—would be preferable from the standpoint of rights. Short, rapid wars would generate less of a Garrison state effect and have a minimal impact on rights. Some domestic surveillance programs might not have happened if the wars were brief. On the other end of the spectrum, longer wars for complete victory would create the J-curve: a highly invested population would have leverage to demand that the government respect their rights. Voters might have eventually demanded reforms: for example, curtailing the FBI’s use of “sneak-and-peak” searches and “National Security Letters” to collect citizens’ private information. Some recent history bears out the hypothetical. For example, after Edward Snowden, an NSA contractor, leaked details of alleged surveillance programs, the US media and population were—momentarily, at least—highly invested in the story. President Barack Obama appointed a panel to study the issue and in 2013 announced a raft of reforms to protect civil liberties.

Rousseau’s main concern is with the impact of rights at home, within the state waging war. But there is another conversation to be had about the impact of war on the rights or liberties of the other side (an omission that might be another quibble with the text). The indefinite and unaccountable authority the US government claimed for itself to unilaterally designate terrorist targets anywhere in the world, at any time, by any means did not affect Americans’ rights (except for a tiny number of American citizens killed by alleged drone attacks because they had apparently defected to the other side). Yet they had a dramatic impact on the liberties of enemy combatants, nearby civilians, and the nations in which they lived.

The kind of drawn-out, inconclusive war that US policy makers were enabled to wage because of voter indifference surely made such tactics more attractive as risk-free, low-cost forms of violence. These types of interventions were attractive to US presidents because they shifted the burden and risk from US voters to foreign nations, minimizing the chance that voters would be affected and the risk that they might start to pay attention. The same interventions exposed swaths of the rest of the world to the possibility of indefinite US military intervention with no end state. That amounts to a violation of people’s right to freedom from fear, freedom to not live in a warzone, or their right to basic public safety.

In other words, presidents chose a form of warfare that helped them evade accountability for their own political convenience at the cost of imposing “endless war” on other countries—because the alternative of helping fix broken societies is too expensive and politically unpopular. Endless war was a feature, not a bug, of US strategy, and its imposition on other nations was immoral. This is an impact of war on rights that would be a fruitful avenue of further research, taking Rousseau’s study as a building block. It is, of course, possible that a highly engaged population might endorse this approach to war anyway because they might not care about the morality of whatever strategy brings them victory or about the rights of the other side—which is another way of saying that voters might not care if the wars they fight are just or not. One hopes that proves not to be the case.